oma-pdf Eval
============

Status: failed
Score: 72/100
Risk: high
Decision: do_not_auto_install
Policy: block
Reason: Permission surface: secrets or environment access, shell or command execution

Install:
npx skills add first-fluke/oh-my-agent --skill oma-pdf

Required checks:
- PASS Task fit: Task wording matches this skill metadata.
- PASS Install path: Install handoff is available.
- PASS Install command safety: standard package or runtime install path
- WARN Trust score: Potentially useful, but at least one trust signal needs human inspection.
- WARN Audit score: Needs review
- WARN Agent safety gate: Sparse or mixed signals. Useful for discovery, but not for autonomous installation.
- PASS License clarity: MIT
- FAIL Permission surface: secrets or environment access, shell or command execution

Warnings:
- Trust score: Potentially useful, but at least one trust signal needs human inspection.
- Audit score: Needs review
- Agent safety gate: Sparse or mixed signals. Useful for discovery, but not for autonomous installation.
- High-risk permission hints: Shell or command execution
- Permission surface may require sandboxing
- Financial research output is not financial advice; require human review before any live investment decision
- The skill runs external code via `uvx opendataloader-pdf` without pinning a specific package version, which introduces supply chain risk if a malicious or compromised version is published.
- The execution protocol excerpt shows a truncated command for the table retry path (`uvx ...`), so that workflow step is incomplete as documented.
- The hybrid OCR server listens on a configurable port, but the documentation does not specify whether it binds to localhost only, which could expose it to local network access.
- There is no explicit guidance on sandboxing or isolating untrusted or potentially malicious PDF inputs before conversion.
- Financial research output is not financial advice; require human review before any live investment decision.
- Quality score needs review

Validation plan:
1. Inspect repository, README/SKILL.md, license, and recent commits before production use.
2. Install in an isolated workspace or sandbox with no production secrets available.
3. Run the smallest representative task and record files touched, commands run, network access, and outputs.
4. Compare the selected skill against at least one alternative when the eval status is review or failed.
5. Promote only after the agent reports a successful verification result and unresolved warnings are accepted.

Do not use when:
- teams that need a vendor-supported SLA
- production agents without a repository review
- The skill runs external code via `uvx opendataloader-pdf` without pinning a specific package version, which introduces supply chain risk if a malicious or compromised version is published.
- High-risk permission hints: Shell or command execution
- Permission surface may require sandboxing
- Financial research output is not financial advice; require human review before any live investment decision
- The execution protocol excerpt shows a truncated command for the table retry path (`uvx ...`), so that workflow step is incomplete as documented.
- The hybrid OCR server listens on a configurable port, but the documentation does not specify whether it binds to localhost only, which could expose it to local network access.

URLs:
- Skill: https://www.openagentskill.com/skills/first-fluke-oma-pdf
- Audit: https://www.openagentskill.com/skills/first-fluke-oma-pdf/audit
- JSON: https://www.openagentskill.com/api/agent/evals?slug=first-fluke-oma-pdf